Empirical Inference Conference Paper 2008

Learning Inverse Dynamics: A Comparison

While it is well-known that model can enhance the control performance in terms of precision or energy efficiency, the practical application has often been limited by the complexities of manually obtaining sufficiently accurate models. In the past, learning has proven a viable alternative to using a combination of rigid-body dynamics and handcrafted approximations of nonlinearities. However, a major open question is what nonparametric learning method is suited best for learning dynamics? Traditionally, locally weighted projection regression (LWPR), has been the standard method as it is capable of online, real-time learning for very complex robots. However, while LWPR has had significant impact on learning in robotics, alternative nonparametric regression methods such as support vector regression (SVR) and Gaussian processes regression (GPR) offer interesting alternatives with fewer open parameters and potentially higher accuracy. In this paper, we evaluate these three alternatives for model learning. Our comparison consists out of the evaluation of learning quality for each regression method using original data from SARCOS robot arm, as well as the robot tracking performance employing learned models. The results show that GPR and SVR achieve a superior learning precision and can be applied for real-time control obtaining higher accuracy. However, for the online learning LWPR presents the better method due to its lower computational requirements.

Author(s): Nguyen-Tuong, D. and Peters, J. and Seeger, M. and Schölkopf, B.
Book Title: Advances in Computational Intelligence and Learning: Proceedings of the European Symposium on Artificial Neural Networks
Journal: Advances in Computational Intelligence and Learning: Proceedings of the European Symposium on Artificial Neural Networks (ESANN 2008)
Pages: 13-18
Year: 2008
Month: April
Day: 0
Editors: M Verleysen
Publisher: d-side
Bibtex Type: Conference Paper (inproceedings)
Address: Evere, Belgium
Event Name: 16th European Symposium on Artificial Neural Networks (ESANN 2008)
Event Place: Bruges, Belgium
Digital: 0
Electronic Archiving: grant_archive
Language: en
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik
Links:

BibTex

@inproceedings{4936,
  title = {Learning Inverse Dynamics: A Comparison},
  journal = {Advances in Computational Intelligence and Learning: Proceedings of the European Symposium on Artificial Neural Networks (ESANN 2008)},
  booktitle = {Advances in Computational Intelligence and Learning: Proceedings of the European Symposium on Artificial Neural Networks},
  abstract = {While it is well-known that model can enhance the control
  performance in terms of precision or energy efficiency, the practical application
  has often been limited by the complexities of manually obtaining
  sufficiently accurate models. In the past, learning has proven a viable alternative
  to using a combination of rigid-body dynamics and handcrafted
  approximations of nonlinearities. However, a major open question is what
  nonparametric learning method is suited best for learning dynamics? Traditionally,
  locally weighted projection regression (LWPR), has been the
  standard method as it is capable of online, real-time learning for very complex
  robots. However, while LWPR has had significant impact on learning
  in robotics, alternative nonparametric regression methods such as support
  vector regression (SVR) and Gaussian processes regression (GPR) offer interesting alternatives with fewer open parameters and potentially higher
  accuracy. In this paper, we evaluate these three alternatives for model
  learning. Our comparison consists out of the evaluation of learning quality
  for each regression method using original data from SARCOS robot
  arm, as well as the robot tracking performance employing learned models.
  The results show that GPR and SVR achieve a superior learning precision
  and can be applied for real-time control obtaining higher accuracy. However,
  for the online learning LWPR presents the better method due to its
  lower computational requirements.},
  pages = {13-18},
  editors = {M Verleysen},
  publisher = {d-side},
  organization = {Max-Planck-Gesellschaft},
  school = {Biologische Kybernetik},
  address = {Evere, Belgium},
  month = apr,
  year = {2008},
  slug = {4936},
  author = {Nguyen-Tuong, D. and Peters, J. and Seeger, M. and Sch{\"o}lkopf, B.},
  month_numeric = {4}
}